jgoy-labs/server-nexe

Local AI server with persistent memory, RAG, and multi-backend inference (MLX / llama.cpp / Ollama). Runs entirely on your machine — zero data sent to external services.